physicsdreams said:
Why does it give me a correct solution?
WolframAlpha is giving a solution in terms of the definition it uses for the function log
10(z). This is not the same as the definition you were given in your class / your text. The domain for the logarithm function as defined for your class is the positive real numbers. Plug your solution, x = -400/99, back into the original equation and you'll get log(-400/99)+log(-4/99)=2. Does that make any sense in terms of what you have been taught?
Sometimes when you solve a problem you will get superfluous answers. You need to learn to recognize that some answers are superfluous. A completely different example: Suppose you have calculated that the distance
d between a pair of points is given by
d2-
d=2. This has two solutions,
d=2 and
d=-1. That second result (
d=-1) is superfluous because distance can never negative. The only solution is
d=2.
Edit
I'll walk through the steps I assume you made to obtained an answer of -400/99.
Note: I am assuming that log(x) means the base 10 logarithm. (You wouldn't get -400/99 if log(x) indicates the natural log.)
Problem statement:
log(x)-log(x+4)=2
Step 1: Use the fact that log(a)-log(b)=log(a/b)
log(x/(x+4))=2
Step 2: Use the fact that log(a)=b is equivalent to a=10
b
x/(x+4)=100
Step 3: Multiply both sides by x+4
x=100(x+4)
Step 4: Solve for x
x=-400/99
Each step introduces the possibility of a superfluous solution. For example, the equivalent of step 3 is used in invalid but creative ways to prove that 1=2. In this case, it is that first step that is problematic. log(a)-log(b)=log(a/b) is valid only if log(a) and log(b) are defined. You should always check your work.